Browse By Day
Browse By Time
Browse By Person
Browse By Room
Browse By Unit
Browse By Session Type
Annual Meeting Housing and Travel
In this study, this conceptual question is empirically investigated through the systematic comparison of “doubly latent” (Marsh et al., 2012; Morin, Marsh, Nagengast, & Scalas, 2014) measurement models that posit differing dimensional structures and school- or student-level specifications of student learning opportunities. Specifically, a large (N = 963) and diverse sample of high-school students, attending nine different high-schools, was analyzed. Results suggest that student perceived learning opportunities are best conceptualized as nine distinct but positively correlated latent factors, and that those factors are best measured at the student, rather than school, level. However, a statistical correction for the school-based clustering in the dataset was required in order to improve the fit of the model, although that “design-based” correction should not change the theoretical interpretation of the model-estimated latent quantities.
After identifying this best fitting measurement model, the model parameters relating to reliability and internal validity were systematically investigated, as well as the latent correlations among the model factors. Moreover, all response-category thresholds, relating to the psychometric functioning of every individual item administered to students, were also examined. Finally, a construct reliability statistic (i.e., Hancock’s H; Gagné & Hancock, 2006; McNeish, 2017) was estimated for each latent quantity estimated in the model. Taken together, these results suggested that the design-based corrected doubly latent measurement model, featuring nine correlated factors at the student level, appeared to be a reliable and valid tool for substantive inferences: including the major inference that student learning opportunities, at least as measured here, occur at the student, rather than school, level. Put another way, perceptions of opportunities for learning among students were so heterogeneous within schools, that the school-level aggregation of these perceptions (perhaps with the goal of evaluating the learning climate of a school) would appear to be invalid.